bio

Type Members

1. class BioPOSPostProcessor extends SentencePostProcessor

   

   Fixes some common POS tagging mistakes in the bio domain (in place)

   Fixes some common POS tagging mistakes in the bio domain (in place)

   Note: this class is used by the CRF-based BioNER to cleanup its training data (from BioCreative 2), through org.clulab.processors.bionlp.BioNLPPOSTaggerPostProcessor. This means that every time there are changes here, the CRF should be retrained. Tell Mihai. User: mihais Date: 9/23/17

2. class BioTokenizerPostProcessor extends TokenizerPostProcessor

   

   Processes tokenization so it suits bio analysis

3. class BioTokenizerPreProcessor extends TokenizerPreProcessor

   

   Preprocesses bio text, including Unicode normalization, and removing figure and table references User: mihais Date: 9/10/17

4. case class ComplexSubToken(text: String, start: Int, length: Int) extends Product with Serializable

   

5. case class PostProcessorToken(word: String, beginPosition: Int, endPosition: Int) extends Product with Serializable